Scope: These ESCMID guidelines address the targeted antibiotic treatment of third-generation cephalosporin-resistant Enterobacterales (3GCephRE) and carbapenem-resistant Gram-negative bacteria, focusing on the effectiveness of individual antibiotics and on combination versus monotherapy.

Methods: An expert panel was convened by ESCMID. A systematic review was performed including randomized controlled trials and observational studies, examining different antibiotic treatment regimens for the targeted treatment of infections caused by the 3GCephRE, carbapenem-resistant Enterobacterales, carbapenem-resistant Pseudomonas aeruginosa and carbapenem-resistant Acinetobacter baumannii. Treatments were classified as head-to-head comparisons between individual antibiotics and between monotherapy and combination therapy regimens, including defined monotherapy and combination regimens only. The primary outcome was all-cause mortality, preferably at 30 days and secondary outcomes included clinical failure, microbiological failure, development of resistance, relapse/recurrence, adverse events and length of hospital stay. The last search of all databases was conducted in December 2019, followed by a focused search for relevant studies up until ECCMID 2021. Data were summarized narratively. The certainty of the evidence for each comparison between antibiotics and between monotherapy and combination therapy regimens was classified by the GRADE recommendations. The strength of the recommendations for or against treatments was classified as strong or conditional (weak).

Recommendations: The guideline panel reviewed the evidence per pathogen, preferably per site of infection, critically appraising the existing studies. Many of the comparisons were addressed in small observational studies at high risk of bias only. Notably, there was very little evidence on the effects of the new, recently approved, β-lactam/β-lactamase inhibitors on infections caused by carbapenem-resistant Gram-negative bacteria. Most recommendations are based on very-low- and low-certainty evidence. A high value was placed on antibiotic stewardship considerations in all recommendations, searching for carbapenem-sparing options for 3GCephRE and limiting the recommendations of the new antibiotics for severe infections, as defined by the sepsis-3 criteria. Research needs are addressed.

范围：这些ESCMID指南涉及第三代耐头孢菌素肠杆菌（3GCephRE）和耐碳青霉烯革兰氏阴性菌的靶向抗生素治疗，重点关注单个抗生素的有效性以及联合治疗与单药治疗的效果。

方法：由ESCMID召集专家小组。对3GCephRE、耐碳青霉烯肠杆菌、耐碳青霉烯绿脓杆菌假单胞菌和耐碳青霉烯鲍曼不动杆菌引起的感染进行了系统评价，包括随机对照试验和观察性研究，研究了不同的抗生素治疗方案。治疗被分类为单独抗生素之间的正面比较以及单一治疗和联合治疗方案之间的正面比较，包括仅定义的单一治疗和联合治疗方案。主要结局是全因死亡率，最好是在30天内，次要结局包括临床失败、微生物失败、耐药性的发展、复发/复发、不良事件和住院时间。最后一次对所有数据库的搜索是在2019年12月进行的，随后是对相关研究的重点搜索，直到2021年ECCMID。对数据进行叙述总结。抗生素之间以及单药治疗和联合治疗方案之间的每次比较的证据的确定性根据GRADE建议进行分类。推荐或反对治疗的强度分为强或有条件的（弱）。

建议：指南小组审查了每个病原体，最好是每个感染部位的证据，批判性地评价了现有的研究。许多比较仅在高偏倚风险的小型观察性研究中进行。值得注意的是，最近批准的新的β-内酰胺/β-内酰胺酶抑制剂对碳青霉烯耐药革兰氏阴性菌引起的感染的影响几乎没有证据。大多数建议都是基于非常低和低确定性的证据。在所有推荐中，高度重视抗生素管理方面的考虑，寻找3GCephRE的碳青霉烯类选择，并限制根据败血症-3标准定义的严重感染的新抗生素的推荐。研究需要得到解决。